News for hackers. Stuff that matters.

A comment on What ChatGPT Can't Do. Parent comment

Re:What ChatGPT Can't Do by layer8 on Saturday December 24, 2022 @10:47PM (#34124040)
That doesn’t explain why it still provides two answers (getting the number of letters right despite tokenization) and also doesn’t notice that they are contradictory.
[ Parent | on HN ]
Re:What ChatGPT Can't Do by 152334H on Sunday December 25, 2022 @03:25AM (#34125392)
Forgive the anthropomorphising, but I find it very similar to how I'd handle someone forcing me to explain two contradictory claims coherently.
[ Parent | on HN ]
Re:What ChatGPT Can't Do by layer8 on Sunday December 25, 2022 @07:05AM (#34126512)
But no one was forcing it to do that, and it doesn’t seem it has any awareness that it could proceed differently.
[ Parent | on HN ]